Recombination between direct repeats has been studied in Penicillium chrysogenum using strain TD7-88 (lys- py+), which contains two inactive copies of the lys2 gene separated by 4.5 kb of DNA (including the pyrG gene) in its genome. Gene conversion leading to products with the lys+ pyr+ phenotype was observed at a frequency of 1 in 3.2x10(3) viable spores. Two types of deletion events giving rise to lys+ pyr- and lys- pyr- phenotypes were obtained with different frequencies. Southern analysis revealed that gene conversion occurs mainly as a result of crossing over events that remove the BamHI frameshift mutation present in one of the repeats. In lys- pyr- recombinants, the deletion events do not affect the frameshift mutation in the BamHI site, while lys+ pyr- recombinants showed repair of the BamHI frameshift mutation and the genotype of the parental non-disrupted strain was restored. In summary, deletion events in P. chrysogenum tend to favor the restoration of the phenotype and genotype characteristic of the parental non-disrupted strain.

A 5.2-kb NotI DNA fragment isolated from a genomic library of Acremonium chrysogenum by hybridization with a probe internal to the Penicillium chrysogenum lys2 gene, was able to complement an alpha-aminoadipate reductase-deficient mutant of P. chrysogenum (lysine auxotroph L-G-). Enzyme assays showed that the alpha-aminoadipate reductase activity was restored in all the transformants tested. The lys2-encoded enzyme catalyzed both the activation and reduction of alpha-aminoadipic acid to its semialdehyde, as shown by reaction of the product with p-dimethylaminobenzaldehyde. The reaction required NADPH, and was not observed in the presence of NADH. Sequence analysis revealed that the gene encodes a protein with relatively high similarity to members of the superfamily of acyladenylate-forming enzymes. The Lys2 protein contained all nine motifs that are conserved in the adenylating domain of this enzyme family, a peptidyl carrier domain, and a reduction domain. In addition, a new NADP-binding motif located at the N-terminus of the reduction domain that may form a Rossmann-like betaalphabeta-fold has been identified and found to be shared by all known Lys2 proteins. The lys2 gene was mapped to chromosome I (2.2 Mb, the smallest chromosome) of A. chrysogenum C10 (the chromosome that contains the "late" cephalosporin cluster) and is transcribed as a monocistronic 4.5-kb mRNA although at relatively low levels compared with the beta-actin gene.

Two strategies have been used for targeted integration at thelys2 locus of Penicillium chrysogenum. In the first strategy the disruption of lys2 was obtained by a single crossing over between the endogenous lys2 and a fragment of the same gene located in an integrative plasmid.lys2-disrupted mutants were obtained with 1.6% efficiency when the lys2 homologous region was 4.9 kb, but no homologous integration was observed with constructions containing a shorter homologous region. Similarly,lys2-disrupted mutants were obtained by a double crossing over (gene replacement) with an efficiency of 0.14% by using two lys2 homologous regions of 4.3 and 3.0 kb flanking thepyrG marker. No homologous recombination was observed when the selectable marker was flanked by short lys2 homologous DNA fragments. The disruption of lys2 was confirmed by Southern blot analysis of three different lysine auxotrophs obtained by a single crossing over or gene replacement. Thelys2-disrupted mutants lacked α-aminoadipate reductase activity (encoded by lys2) and showed specific penicillin yields double those of the parental nondisrupted strain, Wis 54-1255. The α-aminoadipic acid precursor is channelled to penicillin biosynthesis by blocking the lysine biosynthesis branch at the α-aminoadipate reductase level.

The ␣-aminoadipate reductase (␣-AAR) of Penicillium chrysogenum, an enzyme that activates the ␣-aminoadipic acid by forming an ␣-aminoadipyl adenylate and reduces the activated intermediate to ␣-aminoadipic semialdehyde, was purified to homogeneity by immunoaffinity techniques, and the kinetics for ␣-aminoadipic acid, ATP, and NADPH were determined. Sequencing of the N-terminal end confirmed the 10 first amino acids deduced from the nucleotide sequence. Its domain structure has been investigated using limited proteolysis and active site labeling. Trypsin and elastase were used to cleave the multienzyme, and the location of fragments within the

No abstract